EASy Summer 2006Non-Symbolic AI lec 41 Non-Symbolic AI lecture 4 A major difference between Symbolic...

Summer 2006Non-Symbolic AI lec 4 1

Non-Symbolic AI lecture 4Non-Symbolic AI lecture 4Non-Symbolic AI lecture 4Non-Symbolic AI lecture 4

A major difference between Symbolic and Non-Symbolic AI approaches is in modelling, or emulating, Cognition or control – in artificially intelligent machines such as robots.

Symbolic, or Classical, AI tended to think in terms of control being focussed within a central, reasoning brain.

Given a task (for a human or a robot) such as ‘open the door’ or ‘catch the ball’, Symbolic AI assumes that the task can be turned into a set of propositions, using probably logic and maths.

Then this is now a ‘problem to be solved’ using the brain as a computer (… or the computer as a brain !)

Robotics is used for …Robotics is used for …Robotics is used for …Robotics is used for …

… publicising the technical expertise of car companies – the Honda robot

… working out how expressions communicate emotions

… toys

… and for science

-- as a way of understanding how animals and humans work by trying to build artificial ones.

Artificial Life.

Creating Robots in Man’s ImageCreating Robots in Man’s ImageCreating Robots in Man’s ImageCreating Robots in Man’s Image

Whether or not God created Man in His image, it is inevitably the case that Man and Woman create robots in their image.

Puppets, revealing how we (… those in the robot/cognitive science or philosophy business) really think of ourselves.

Doing ‘Philosophy of Mind’ with robots has one enormous disadvantage over conventional philosophy … …

… … you cannot fudge things, or appeal to magic!

Brains and BodiesBrains and BodiesBrains and BodiesBrains and Bodies

There is a traditional view that all the intelligence of a creature is in some rational brain – maybe like a computer – and the body is just ‘an afterthought’.

Here is an 8-legged walking robot like this – with an “artificially evolved brain” sitting inside the onboard computer.

CognitionCognitionCognitionCognition

21st Century scientific human cognition is different from that ofhumans 3000 years ago is different from that ofour ancestors of 2 billion years ago is different from that ofour descendants of 2 billion years later (... if there will be any ...)

DescartesDescartesDescartesDescartes

Much of classical AI can be traced back to Descartes (early 17thC)

Dualism -- the separation of the mental and the physical. Cartesian objectivity:

"there just is a way the world is, independent of any observer. The scientist is a spectator from outside, a God's-eye view"

"The world is physical, knowledge is mental (something different)"

The view from outsideThe view from outsideThe view from outsideThe view from outside

Classical AIClassical AIClassical AIClassical AI

When building robots, this gives Classical AI approach where the robot is a scientist-spectator, seeking information from outside.

"SMPA" -- so-called by Brooks (1999)S senseM modelP planA action

sensory inputs model or representation

The model is 'computed' from the sensory inputs.

But what is the computer metaphor?

Computing a modelComputing a modelComputing a modelComputing a model

The Computer metaphorThe Computer metaphorThe Computer metaphorThe Computer metaphor

A Turing machine is a formal way of carrying out an algorithm -- a list of explicit instructions.

BUT beware of a simple confusion:-

When the astronomer calculates where the moon will be at 12:00 noon on May 1st, she carries out computations. She is a scientist-spectator.

But the moon does not carry out computations -- it 'just moves' in a deterministic way.

Classical AI confusionClassical AI confusionClassical AI confusionClassical AI confusion

The Classical AI approach tends to confuse these two -- tends to (mistakenly) think that "the brain does computations".

To clarify: we can use a computer to simulate (predict) the movement of the moon -- even to control a model planetary system.

Similarly we can use a computer to simulate (predict) the dynamics of a nervous system -- even to control a robot with a model 'brain'

-- but this does not mean that the “brain computes” !

‘‘Reasoning all the way down’Reasoning all the way down’‘‘Reasoning all the way down’Reasoning all the way down’

The Classical AI approach, obsessed with reasoning and computing, assumed that even something as simple as walking across the room, maintaining one’s balance, required reasoning and computation … …

… … “Sense Model Plan Action” …

… … Brain controlling muscles

But look at this ---

Passive Dynamic WalkingPassive Dynamic WalkingPassive Dynamic WalkingPassive Dynamic Walking

‘Natural walking behaviour', stable to small perturbations, can emerge from 'all body and no brain' !It is the dynamics that count, whether the dynamics arise with or without a coupled nervous system.

Dan Jung’s walker movie www.msc.cornell.edu/~ruinalab/pdw.html"Passive Dynamic Walking", from Tad McGeer

Walking without a nervous systemWalking without a nervous systemWalking without a nervous systemWalking without a nervous system

Alternatives to the Classical ApproachAlternatives to the Classical ApproachAlternatives to the Classical ApproachAlternatives to the Classical Approach

There are different philosophical perspectives such as those of Heidegger / Merleau-Ponty / Wittgenstein that might affect the way in which one designs robots.

These are difficult people to read, and they say little or nothing about robots !

Nevertheless, they offer a different perspective which has recently been crucially important

Other sourcesOther sourcesOther sourcesOther sources

Brooks 1999. Cambrian IntelligenceDreyfus 1972. What Computers Can't DoWinograd and Flores 1986. Understanding Computers and Cognition.Pfeifer and Scheier 1999. Understanding IntelligenceMaturana and Varela 1987. The Tree of Knowledge

Situatedness and Embodiment

The Dynamical Systems view of Cognition

Heidegger …Heidegger …Heidegger …Heidegger …

…rejects the simplistic objective view, that the "objective physical world is the primary reality that we can be sure of"

He also rejects the opposite idealistic/subjective view that "our thoughts are the primary reality"

The primary reality is our everyday practical lived experience, as we reach for the coffee or switch on the light

This is more fundamental than detached theoretical reflection.

Reasoning only came later …Reasoning only came later …Reasoning only came later …Reasoning only came later …

This actually makes sense from a Darwinian evolutionary perspective (though Heidegger would not say this)

-- our human language / reasoning powers arrived only 'recently‘ (last few 10,000 years, 100,000s ?)

From a phylogenetic and ontogenetic view, we are organisms/animals first -- thinking humans only later.

Our unreflective tool-using is primary -- only when something goes wrong do we need to switch into 'reflective' mode.

What comes first?What comes first?What comes first?What comes first?

Any lessons for robotics?Any lessons for robotics?Any lessons for robotics?Any lessons for robotics?

This is true (Wittgenstein suggests) even for language skills:

"In general we don't use language according to strict rules -- it hasn't been taught us by means of strict rules either"

What lessons for robots from these alternative views? At first sight, they are negative and unhelpful !

For everyday robot actions this implies we should do without planning, without the computational model, without internal representations ... .... but what should we do instead ?

Dynamic skills all the way up?Dynamic skills all the way up?Dynamic skills all the way up?Dynamic skills all the way up?

Perhaps rather than ‘Reasoning all the way down’ …

… we should think in terms of ‘Dynamic skills all the way up’

Two initial lessons -- cognition isTwo initial lessons -- cognition isTwo initial lessons -- cognition isTwo initial lessons -- cognition is

Situated: a robot or human is always already in some situation, rather than observing from outside.

Embodied: a robot or human is a perceiving body, rather than a disembodied intelligence that happensto have sensors.

EASy ...animals are endowed with nervous systems whose

dynamics are such that, when coupled with the dynamics of their bodies and environments, these animals can engage in the patterns of behavior necessary for their survival"

Beer & Gallagher 1992.

The Dynamical Systems view of CognitionThe Dynamical Systems view of CognitionThe Dynamical Systems view of CognitionThe Dynamical Systems view of Cognition

A Crucial DifferenceA Crucial DifferenceA Crucial DifferenceA Crucial Difference

What is one crucial difference between the Classical AI approach and the Dynamical Systems approach ?

Classical AI and computational approaches do not take account of time -- 'life as a series of snapshots

Dynamical Systems approach -- time is central, 'life as process'

EASy How can you design Dynamical Nervous Systems?

Brooks' Subsumption architecture is one way.

Evolutionary Robotics is another.

(Something crudely like the way we humans were

designed !)

Subsumption architecture (1)Subsumption architecture (1)Subsumption architecture (1)Subsumption architecture (1)

(1a)(1a)(1a)(1a)

Traditional decomposition of a mobile robot control system into functional modules

Brooks’ alternativeBrooks’ alternativeBrooks’ alternativeBrooks’ alternative

Brooks’ alternative is in terms of many individual and largely separate behaviours – where any one behaviour is generated by a pathway in the ‘brain’ or control system all the way from Sensors to Motors.

No Central Model, or Central Planning system.

(1b)(1b)(1b)(1b)

Decomposition of a mobile robot control system based on task-achieving behaviors

Subsumption architecture (2)Subsumption architecture (2)Subsumption architecture (2)Subsumption architecture (2)

(2a)(2a)(2a)(2a)

Level 3

Level 2

Level 1

Level 0SENSORS

ACTUATORSControl is layered with higher levels subsuming control of lower layers when they wish to take control.

SubsumingSubsumingSubsumingSubsuming

‘Subsume’ means to take over or replace the output from a ‘lower layer’.

The 2 kinds of interactions between layers are

1. Subsuming

2. Inhibiting

Generally only ‘higher’ layers interfere with lower, and to a relatively small extent – this assists with an incremental design approach.

EASy Summer 2006Non-Symbolic AI lec 41 Non-Symbolic AI lecture 4 A major difference between Symbolic...

Documents

Developmental trajectories of symbolic and non- symbolic … · 2008-03-12 · Developmental trajectories of symbolic and non-symbolic number processing: behavioral and brain-imaging

Lecture 1 Introducing Symbolic AI - COMP24412: Symbolic AIsyllabus.cs.manchester.ac.uk/.../lecture1.pdf · The main AI disciplines required to pass the so-called total Turing Test

Automated Edge Symbolic AI Asset Intelligenceinscription.firstconnection.fr/images/files/... · Embedded Predictive Objects PO in IIoT platforms Micro Predictive Objects Paris & Montpellier

EASy Summer 2006Non-Symbolic AI lecture 51 We shall look at 2 alternative non-symbolic AI approaches to robotics Subsumption Architecture Evolutionary

Non Non-Symbolic AI lecture 8-Symbolic AI lecture 8users.sussex.ac.uk/~inmanh/non-symb/nsai2005/NS08/lec8.pdf · Non-Symbolic AI lec 8 3Summer 2005 EASy Differentiable sigmoidDifferentiable

Fractal Symbolic Analysispingali/CS380C/2016/papers/fractal.pdf · Fractal Symbolic Analysis † 779 Fig. 2. Overview of fractal symbolic analysis. an approximate symbolic analysis

EASy Summer 2006Non-Symbolic AI lecture 31 More on Evolutionary Algorithms Last lecture discussed encoding real numbers as bits on a genotype (either binary

Comp24412: Symbolic AI - Lecture 2: Prolog IIsyllabus.cs.manchester.ac.uk/.../COMP24412/lectures/large/prolog2.pdfProblem: write a program to compute ... SWI Prolog allows both.

Logics and symbolic AI: Knowledge representation and …...Logics and symbolic AI: Knowledge representation and reasoning Isabelle Bloch, Natalia D az Rodr guez LTCI, T el ecom Paris

SenticNet 6: Ensemble Application of Symbolic and Subsymbolic … · SenticNet 6: Ensemble Application of Symbolic and Subsymbolic AI for Sentiment Analysis Erik Cambria Nanyang Technological

EASy Summer 2005Non-Symbolic AI lec 121 Non-Symbolic AI lecture 12 Evolution of Communication In particular, 2 papers from Proc of Artificial Life II ed

From Symbolic Violence to Symbolic Reparation

EASy Summer 2006Non-Symbolic AI lec 61 Non-Symbolic AI lecture 6 Last lecture we looked at non-symbolic approaches to robotics – new methods of approaching

Neural Networks AI – Week 23 Sub-symbolic AI Multi-Layer Neural Networks Lee McCluskey, room 3/10 Email lee@hud.ac.uklee@hud.ac.uk

EASy Summer 2006Non-Symbolic AI lecture 11 Non-Symbolic AI – Summer 2006 Lecturer: Inman HarveyPEV2 rm 5C12 x8431 inmanh@susx.ac.uk

Integration of symbolic and connectionist AI techniques in ...ment of wastewater treatment plants using formal veriﬁcation. In International Symposium on Sanitary and Environmental

Comp24412: Symbolic AI - Lecture 2: Prolog II

EASy Summer 2006Non-Symbolic AI lec 111 Non-Symbolic AI lecture 11 Evolution of Communication In particular, 2 papers from Proc of Artificial Life II ed

Symbolic Execution...Symbolic execution •King, CACM 1976. •Key idea: generalize testing by using unknown symbolic variables in evaluation •Symbolic executor executes program,

AI – Week 23 Sub-symbolic AI Multi-Layer Neural Networks